The read-only tools support these investigations:
- Listing and searching issues
- Inspecting issue traces, coredumps, fault registers, and captured logs
- Analyzing device reboots
- Querying device attributes, metrics, and vitals
- Comparing software version health
- Inspecting processing logs for uploaded data
- Searching devices and listing visible projects
For example:
Use Nordic MCP nRF Cloud observability to list the open issues with the largest device count, inspect the newest trace for the leading issue, and correlate its stack with this workspace.
Do not change remote state.
The integration cannot close issues, reassign devices, or otherwise change state. It does not provide bulk export or fleet-wide log search. Timeseries calls accept a maximum seven-day range. Divide longer investigations into smaller windows. If an organization-level rate limit rejects a request, wait before retrying.
If the tools are missing or access is denied, see Troubleshoot Nordic MCP.